EchoZero

A trading agent can have a personality, several strategies, a public profile, and access to allocated funds. Those layers need distinct controls. Changing an agent’s identity should not be confused with changing its trading rules or the way a strategy executes.

We separated the agent from its strategies while keeping their relationship visible. People can discover agents, inspect their activity, create their own rules, and decide which actions require approval. Chat sits within that structure rather than replacing it.

Strategies can start manually, from a template, or through a conversation. The resulting logic combines conditions and actions, including price, volume, market, and social signals.

We designed a structured rule builder around grouped conditions and explicit relationships such as when, if, and then. Review brings the agent and its strategy together before activation. People can inspect what the system is being asked to do instead of relying on a conversational summary alone.

Full Trade, Test Trade, and Alerts represent different relationships with the system. Full Trade executes with real funds. Test Trade uses virtual funds. Alerts asks for approval before execution.

The interface distinguishes test trading visually as well as by name, shifting the app’s accent from orange to pink. Strategy controls support pausing and editing, while a stop‑all action gives people a direct way to interrupt execution across their strategies.

A wallet total alone does not explain which funds are committed. We separated total, in‑use, and available balances so people can distinguish their overall position from the funds available for another action.

Portfolio and strategy views connect balances with activity, history, and performance. Agent profiles add the context needed to inspect a strategy before using it. The design keeps discovery and execution connected to information people can review.

People can ask questions, move into an agent‑specific conversation, or start building a strategy through chat. The selected agent and its context remain part of that interaction, so switching conversations does not mean losing track of who is being asked to act.

Explore, chat, earnings, and inbox views support different reasons to return. Together they connect strategy discovery, ongoing questions, creator activity, and updates from the system.